HMRF_EM_image

tiff图像的读取与处理，并且说明清新，有很大的帮助。(The reading and processing of the tiff image, and the description of the fresh, is of great help.)

采集RGB-D深度图像生成点云矩阵，属于RGB-D SLAM的前端处理过程(Collecting depth image to generate point cloud)

说明：  生成显著图方法，一种基于空间频域分析的显著性分析算法（Spectral Residual，SR算法）的matlab实现代码(Matlab code generated saliency map method based on the spatial frequency domain analysis of significant analysis algorithm (Spectral Residual, SR algorithm))

mser算法（Maximally Stable Extremal Regions）代码，非文章作者写的，但是还不错，matlab实现。(mser algorithm (Maximally Stable Extremal Regions) code, non-article the authors write, but also good, matlab realize.)

提出一种基于特征点的运动场景图像拼接算法。基于相似变换消除选取在运动物体上的特征点，并使用RANSAC算法进行外点的剔除。在融合阶段，根据齐次变换矩阵求得对2幅图像的差异图像，并执行区域增长算法。从而分割出运动区域。使用一种分段映射算法生成全景图像。实验证明了算法的有效性。(A feature-based algorithm for image mosaics of 8c宅Iles with moving objects is pI．oposcd．The algorithm eliminates the feature points which arc selected in moving objects using similarity transform，and the outliers are eliminated using RANSAC algorithm．In the image blend phase，a difference of two images is created using homogeneous transformation matrix，and a region growth algorithm is implemented SO as to segment the movmg object region．A piecewise mapping algorithm for generating panoramic images is used．Experimental results show the effectiveness of the algorithm．)

实现文章“Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries”中的降噪算法(Realize the article " Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries" in the noise reduction algorithm)

Image corrosion, swelling, binarization processing, image binary segmentation procedure matlab gui

对灰度图像建立高斯金字塔，可以调整核的参数。(Establish Gaussian pyramid for gray-scale images to , you can adjust the reduce parameters.)

说明：  本代码总结了信息熵、条件信息熵、PSNR、SSIM等图像评价指标(This code summarizes information entropy, conditional information entropy, PSNR, SSIM and other image evaluation indexes)

适合初学小波的同志，小波图像增强原理底层。严格高低频分解重构，没有使用小波工具库。(Suitable for beginners wavelet comrades, the underlying principle of wavelet image enhancement. Strict high-low-frequency decomposition of reconstruction, did not use the wavelet tool library.)